Antibodies are defense proteins produced by the vertebrate adaptive immune system for the purposes of binding and targeting for clearance of a diverse range of bacteria, viruses, and other foreign molecules (collectively referred to as antigens) (see, for e.g., Abbas et al. (1997), Cellular and Molecular Immunology, 3rd Ed., Chapter 3, pp. 37-65). As a result of their ability to bind target antigens selectively and with high affinity, antibodies are useful tools for protein purification, cell sorting, diagnostics, and therapeutics.
Conventional antibody production has involved the immunization of animals (i.e., mice) with a target antigen, such as a virus, bacteria, foreign protein, or other molecule. The immunized mice produce on the order of 104-105 antibody secreting cells (ASCs), each with the capacity to produce a unique (monoclonal) antibody specific to the target antigen (see, for e.g., Poulson et al. (1997), J. Immunol. 179: 3841-3850; and Babcock et al. (1996), Proc. Natl. Acad. Sci. USA 93: 7843-7848).
The ASCs are then harvested from the immunized animals and screened in order to select which cells are producing antibodies of desired affinity and selectivity to the target antigen. Since single ASCs do not produce antibodies in sufficiently large quantities for binding affinity measurements, each ASC is clonally expanded. Primary ASCs do not grow efficiently in laboratory tissue cultures; thus, clonal expansion may be achieved by fusing ASCs to murine myeloma (cancer) cells to produce immortalized, antibody-secreting (hybridoma) cells (see, for e.g., Kohler, G. and Milstein, C. (1975), Nature 256: 495-497). Using this method, expansion of each successfully created hybridoma then produces a monoclonal antibody in sufficiently high concentrations to measure its affinity and selectivity to a target antigen.
It has been recognized that a limitation of hybridoma technology is the low efficiency of the fusion process. For example, whereas an immune response may produce on the order of 104-105 antibody secreting cells, a typical fusion will yield less than 100 viable hybridomas. (see, for e.g., Kohler, G. and Milstein, C. (1975), Nature 256: 495-497; Karpas et al. (2001), Proc. Natl. Acad. Sci. USA 98: 1799-1804; and Spieker-Polet et al. (1995), Proc. Natl. Acad. Sci. USA 92: 9348-9352). Therefore, fusions from hundreds to thousands of animals are required to fully sample the diversity of antibodies produced in an immune response, making the hybridoma approach both time-consuming and expensive. Attempts to circumvent hybridoma generation by immortalizing antibody-producing cells using viral transformations have resulted in modest gains in the efficiency of ASC immortalization. However, these approaches still require costly and time-consuming clonal expansion in order to produce sufficient quantities of monoclonal antibodies to screen for affinity and selectivity to target antigens (see for e.g., Pasqualini, R. and Arap, W. (2004), Proc. Natl. Acad. Sci. USA 101: 257-259; Lanzavecchia et al. (2007), Current Opinion in Biotechnology 18: 523-528; and Traggiai et al. (2004), Nat Med 10: 871-875).
Devices have been developed to estimate the equilibrium dissociation constants of antibodies secreted from single antibody-secreting cells (Story, C. M. et al. Proc. Natl. Acad. Sci. U.S.A. (2008) 105(46):17902-17907; and Jin, A. et al. Nat. Med. (2009) 15(9):1088-1092), but do not measure antibody-antigen binding kinetics using antibodies secreted from single cells.